In Poverty in America: Trends and Explanations, Hoynes, Page, and Stevens research trends in the US poverty rate over the past four decades, and examine the extent to which changes in labor market opportunities, government programs and the population’s demographic composition can explain these trends. This essay reveals that economic indicators are not as helpful in predicting poverty rates as they once were, and that changes in the female labor supply should have reduced poverty, but instead it had little effect because of changes in the number of single mothers that are heads of households.
In the 1960s the poverty rate was cut in half, declining from 22.4% in 1959 to 12.1% in 1969. In the next two decades there was a gradual increase in the poverty rate as the 11.8% average during the 1970s rose to 13.8% during the 1980s. With great economic expansion during 1990s the average poverty rate declined, and as of 2003 was 12.5%. Because there are different causes of poverty among the elderly and non-elderly, and a need to limit the focus of the paper, the rest of the study deals with the poverty rates of the non-elderly poor. The trends of the non-elderly were found to follow the overall trends very closely. Also, all observations deal with the absolute definition of poverty, meaning to be considered poor a family’s, or individual’s, income must be below a certain threshold. Some of the major trends pointed out among the non-elderly poverty rates in 2003 included:
Hoynes, Page, and Stevens next evaluate four factors that have an influence on the poverty rate:
They first examine how labor market opportunities, specifically, growth, inequality and macroeconomic cycles effect the poverty rate. Regional data is used, as oppose to aggregate, to examine the regional variation in the poverty rate based on changes in the labor market factors. The data is broken up into nine different census divisions. The regression analysis shows that poverty rates were significantly higher when unemployment rates were greater, there were lower median wages, and there was higher inequality. An increase in unemployment by 1% increased the poverty rate between .4 and .7%, while a 10% increase in median wage decreased it by about 2%. To measure the effect a more unequal wage distribution (inequality) has on regional poverty rates, a ratio of the wages earned by those in the 50th percentile to the wages of those in the lower 20th percentile was used. Based on this measurement it was estimated that a 10% increase in this ratio leads to a 2.5% increase in the poverty rate. These effects are then examined over three different periods 1967-1979, 1980-1989, and 1990-2003. Splitting the data in this fashion showed the impact of each factor weakening significantly after the first period. It was also observed that when holding two of the factors constant, the poverty rate could be estimated based on the changes of the third factor. In each case there was an over estimate of the true poverty rate, implying that there is some other factor influencing the poverty rate not accounted for. The authors posit that this could be the influence of the fact that more women are entering the workforce. To explore this idea further, a transition is made to examine how changes in typical family structure affect the poverty rate.
Because poverty rates for households headed by women are 3 to 4 times greater than the rest of the population, changes in the distribution of family types can have a big impact on the level of poverty in the US. Statistics show that the number of individuals living in households headed by single women has doubled since 1967, from about 6% to 12%. Examining family structure alone would lead to a predicted increase in poverty rate from the 13.3% in 1967 to 17% in 2003. However, the actual poverty level now is lower than the starting value. This is most likely due to the fact that even though there are more single mothers, there has also been a significant increase in the number of women in the workforce. The fact that there are more single mothers and the fact that more women are in the workforce are pulling the poverty rate in opposite directions.
Finally, the other factors, changes in the number of immigrants coming to the US and changes in the generosity of anti-poverty programs, were examined and both found to not play an important role in influencing the poverty rate.
After reading I have to agree with the authors when they note that changes in poverty rates reflect a complex combination of changes in demographics and changes in labor market conditions. Much more research is needed to learn more about how labor market opportunities and other factors are interacting with each other to influence the poverty rate. Because of this the authors don’t seem to be able to provide us with a solid reason as to why, in the midst of booming economic growth, the poverty rate remains steady, or why changes in the factors influencing poverty rate, after 1980, don’t seem to impact the poverty rate as strongly. The authors do a good job in showing empirically how some factors affect the poverty rate and how their influence on the poverty rate has changed over time, but no real explanation as to why is provided. The only thing discussed was the fact that more women are entering the work force, and that this could be a reason poverty rates aren’t as high as changes in labor market factors would predict. But this is negated by the fact that there are more single women as heads of homes. There is a missing factor. Probably many. Which makes the question of what influences poverty rate a very difficult one to answer, and support empirically.